Kernel panic and system frozen on DL385 G6

[Allen Huang_1]

We have some issues with a few HP ProLiant DL385 G6 dual Six-Core AMD Opteron 2427 @2.20GHz servers recently:

We got errors like:
System1:
Kernel panic – not syncing: nmi watchdog
Kernel panic => reset cleared it

System2:
System frozen
Kernel panic – not syncing: Fatal exception

We recently purchased these systems and they are all running CentOS 5.2

How can we make these systems more stable?

[Steven E. Protter]

Shalom,

Suggestion 1. CentOS 5 update 4.

Rarely have I encountered a RHEL release that was more bullet proof.

Install HP's PSP package and run monitors.

Boot to NVRAM hardware utils and test all CPU's and memory.

SEP

[Allen Huang_1]

Thank you Steven for your suggestions.

[loco_vikide · Accepted solution]

5.2 does not support DL385 Six-Core AMD. You must upgrade to at least 5.3 or higher. I would suggest 5.4 or wait for 5.5 to become GA. 5.5 beta has been released.

Cheers

[Allen Huang_1]

Great info and advice. Thank you!